SCALING IN HEAT TRANSFER || MODULE 3 || EC 465 || KTU MEMS || LECTURE 23
1 Video View·Oct 6, 2023
FOR FULL CLASS FOR MEMS: https://www.youtube.com/playlist?list=PL2UV2EJdMQmiqKJXYU56KbT2APawSe7tr
Hi all, today we are discussing scaling in heat transfer.
So heat transmission in microsystem is in the modes of conduction and convention.
So what do you mean by conduction? Conduction is heat conduction is also called heat diffusion.
Is the direct microscopic exchange of kinetic energy of particles
through the boundary between two systems that is called heat conduction and heat convention occurs
when bulk flow of a fluid or gas or liquid carries heat along with the flow of matter in the fluid.
But in some special cases heat is also transferred by radiation such as laser treatments involved in
some manufacturing techniques for microsystems. There is heat transfer occur in the form of
radiation. So we can discuss scaling in heat transfer in detail.
So first scaling of heat flux. Heat conduction in solid is governed by Fourier law.
According to Fourier law qx is equal to minus k into dou T of x, y, z, T divided by dou x where
qx is the heat flux along the x-axis and k is the thermal conductivity of the solid
and T of x, y, z, T is the temperature field. So rate of heat conduction can be calculated by
the equation rate of heat conduction q is equal to small q into A where A is the area.
So which is equal to minus k into dou T of x, y, z divided by x, y, z, T divided by dou x can be
denoted by delta T divided by delta x. So we can write q as minus k into delta T divided by delta
x into A. For solids in meso scale and micro scale q is proportional to A into
minus k into A into delta T divided by delta x is the equation. So converting into our scale
linear form so we can write it as q is proportional to
L square instead of A we can write L square instead of delta x we can write divided by L.
Other terms are not significant in the case of scaling. So q is proportional to L square into
L raised to minus 1 which is equal to L raised to 1. This means reduction in size leads to
decrease of total heat flow. Next is scaling in submicrometer regime.
In submicrometer regime the thermal conductivity can be calculated as k is equal to
k is equal to k main k is the thermal conductivity k is equal to 1 by 3 into C V lambda
where C is the specific heat, V is the molecular velocity and lambda is the average mean free path
which is proportional to L raised to 1. Here in the case of submicrometer regime k is
proportional to L raised to 1. Thus q is proportional to L raised to 1 into L raised
to 1 that is L square here q is proportional to L square that is that means from this equation
we know that q is proportional to L raised to 1 we get q is proportional to L raised to 1 here
we are neglecting k and delta t divided by delta x only considering k and delta t
here we are only considering a and delta x in the case of meso and micro scales.
But when we are considering submicrometer regime here k is proportional to L raised to 1.
So we have to write here L raised to 1 k for k we have to write L raised to 1 and rest of the
time we will get L raised to 1 q is in the previous discussion we get L raised to 1 as the
scaling value. So here in the case of submicrometer regime q is proportional to instead of k L raised
to 1 and rest k into delta t divided by delta x we write L raised to 1. So L raised to 1 into
L raised to 1 that is L square that means a reduction in size of 10 would lead to reduction
of total heat flow by 100 which is in the case of scaling in submicrometer regime.
Thus we are discussing scaling in effect of heat conduction
in solids of meso and micro scales. A dimensionless number called Fourier number
F0 is used to determine the time increments in transient heat conduction analysis.
So the Fourier number F0 is alpha t divided by L square. Here alpha is the thermal diffusivity
of the material and t is the time for heat to flow across the characteristic length L.
So here from this equation we can write t is equal to t here t is equal to F0 into L square
divided by alpha. Here F0 and alpha are constants so we only we have to here scaling will affect
the characteristic length L. So replacing with our linear value so t is proportional to L square.
Next is scaling in heat convention in previous we are discussed about scaling methods in the
case of heat conduction. Next we are going to discuss scaling in heat convention.
Heat transfer in fluid is in the mode of convention according to a law based on
Newton's cooling law. From this law we can write q is equal to q is equal to q means
the total heat flow between two plates that is q is equal to q into A.
Here q is the heat flux and A is the area cross sectional area for the heat to flow.
From this we can write q A is equal to q is equal to h into delta t so we can write
h A delta t q is equal to h A delta t here h depends on the primarily on the fluid velocity
which does not play any significant role in scaling of the heat flow. So in this case
the heat transfer is entirely depend on the area. So we can write in
meso and micro regimes q is proportional to A that is q is proportional to A so area
in our scaling area is corresponds to L square. So here q is proportional to L square.
For the cases in which gases pass in narrow channels at submicrometer scale.
The classical heat transfer theory is based on the continuum fluid breakdown. The seemingly
convective heat transfer has in fact become conduction of heat among the gas molecules
as the effect of boundary layer becomes a dominant factor.
Here in this figure h is here the height the gap between the two plates the gap h is less than
7 lambda where lambda is equal to 65 nanometer for gases 65 nanometer for gases
and 1.3 micrometer for the liquids. So the equation for lambda is proportional to 1 by rho
and k is equal to 1 by 3 into c into v lambda we know that we studied this equation.
And the equation for this v is v is equal to square root of 8 k T divided by pi m. Here T
is the mean temperature of the gas and m is the molecular weight. Molecular weight of the gas.
So effective heat flux in the case here the effective heat flux q effective is equal to k
into delta T divided by h plus 2 epsilon where delta T is the temperature difference between
the two plates and epsilon depends on the gas entrapped between two plates.
For two plates 2.4 lambda less than epsilon less than 2.9 lambda that means the value of lambda
is in between value of epsilon in between 2.4 lambda and 2.9 lambda for air. In the case of
oxygen nitrogen and carbon dioxide methane and helium the value is in between the value of
epsilon in between 2.4 lambda and 2.9 lambda. Next the value of epsilon is 11.7 lambda
with h greater than 7 lambda for hydrogen. That's the end of today's session. Thank you.

